Advancements in technology contribute to an increased vulnerability to cyberattacks, with Distributed Denial of Service (DDoS) attacks being a prominent threat. Attackers overwhelm network servers with excessive data, hindering legitimate users from accessing them. Software Defined Networking (SDN) is particularly susceptible due to its centralized architecture, making it a prime target for DDoS attacks aimed at the control planes. As cloud computing has grown rapidly, software-defined networks have been developed to provide dynamic management and enhanced performance. Several security concerns are growing, especially as DDoS attacks and malicious actors become more interested in SDN controllers. Many researchers have proposed detecting DDoS attacks. Due to their unqualified features and non-realistic data sets, these approaches have high false positive rates and low accuracy. As a result, SDN controllers can be protected against DDoS attacks using deep learning algorithms (DL). Furthermore, the suggested method comprises three phases: The process involves pre-processing the data, selecting significant features for DDoS detection based on correlation, and utilizing Deep Neural Networks (DNNs) for the detection. In order to evaluate the efficiency of the method proposed, we employ a benchmarking dataset to evaluate the false positive rate as well as detectability, with the traditional assessment indicators. In this paper, we propose a deep learning method for detection of DDoS attacks called DNNADSC, which is the first anomaly detection method based on deep neural network for DDoS attacks. The method proposed efficaciously recognizes DDoS attacks, with the detection rate of 99.39%, with a precision of 97.41% with a false-positive rate (FPR) that is 0.0665 with the F1 measure of 99.32%.

The security of public key cryptosystems has become a major concern due to recent developments in the field of quantum computing. Despite efforts to enhance defenses against quantum attacks, current methods are impractical due to safety and efficacy concerns. A recent study explores hash-based digital signature methods and evaluates their effectiveness using Merkle trees. Furthermore, novel approaches based on Verkle trees and vector commitments have been studied to reduce quantum threats. 
First, we introduce a post-quantum digital signature system that combines vector commitments based on lattices with Verkle trees. This architecture optimizes traditional Merkle tree architecture by preserving resistance to quantum attacks while improving cryptographic proofs. Second, in order to ensure secure initial seed generation without sacrificing operational viability, we create a hybrid random number generation framework that combines quantum random number generation (QRNG) with pseudorandom approaches. We provide a detailed analysis of generating random numbers in our article, which makes it easier to build a post quantum cryptosystem that uses our generator to provide initial random values. Our system is notable for its robust security against quantum threats, speed, and efficiency.

Modern communication networks face significant challenges due to the constant growth of traffic volumes and the need to effectively manage network resources. Standard routing protocols based on finding a single shortest path can lead to uneven load distribution and limit the overall throughput. One of the promising approaches to solving these problems is multi-path routing, which allows the use of multiple available paths for data transmission. In this paper, we propose a modification of the Dijkstra’s algorithm that extends the classical approach to find a set of optimal routes in a single algorithm run. The developed modification allows forming not only the main tree of shortest paths, but also additional trees of alternative routes, saving them based on certain conditions.

Wireless sensor networks (WSNs) play a critical role in various applications such as environmental monitoring, healthcare, and industrial automation. The Distributed Energy-Efficient Clustering (DEEC) algorithm has been widely used for efficient data gathering and energy management in WSNs. However, the selection of cluster heads (CHs) in DEEC and data aggregation remain challenging tasks that significantly impact the performance and lifetime of the network. In this paper, we propose a novel approach for cluster head selection in the Distributed Energy-Efficient Clustering (DEEC) algorithm, utilizing the Particle Swarm Optimization-Gravitational Search Algorithm (PSO-GSA). Our approach enhances the CH selection process in DEEC by leveraging the strengths of both PSO and GSA, resulting in more optimal CH selection considering energy efficiency and network coverage. Furthermore, we employ fuzzy logic for data aggregation, which improves the accuracy and efficiency of sensor data aggregation. Our proposed hybrid approach, combining PSO-GSA for CH selection and fuzzy logic for data aggregation, is unique and original, and contributes to the advancement of WSNs and optimization techniques. Through extensive simulations and analysis, we demonstrate the effectiveness and superiority of our proposed approach over existing methods. This paper presents a significant advancement in WSN optimization techniques, promising enhanced energy efficiency and robustness in practical applications. Our approach achieves up to 36.66% and 60.45% increase in first node dead compared to DEEC in DL-DEEC with DA, highlighting its superior performance in prolonging network lifetime.

Vehicular Ad hoc Networks (VANETs) are vital for efficient and secure vehicle-to-infrastructure communication in intelligent transportation systems. sybil attacks, where malicious entities adopt multiple identities, are a major security concern in VANETs. Detecting and mitigating these attacks is crucial for ensuring communication reliability and trust. This article focuses on detecting sybil attacks in Vehicle-to-Vehicle (V2V) communication by using a novel mechanism that characterizes the wireless channel through Received Signal Strength Indicator (RSSI) and angular spread in both azimuth and elevation planes. By incorporating angular spread alongside RSSI, the proposed mechanism offers more accurate and robust detection, particularly in dense vehicle environments. Utilizing a precise wireless channel model based on ray tracing statistics, the approach outperforms traditional RSSI-based methods. Experimental results confirm the enhanced accuracy and reliability of the proposed mechanism for detecting sybil attacks in V2V communication scenarios.

The purpose of this work is to develop an algorithm and a method for identifying key nodes involved in the dissemination of criminal information within social networks. This study focuses on social network analysis (SNA) metrics that facilitate the detection of influential actors in organized groups, particularly activists who serve as primary disseminators of criminal content. The research objects include both the textual content and metadata of users on social media platforms such as "Vkontakte" and "YouTube." To achieve this goal, an algorithm for detecting nodes that distribute criminal information has been developed. A conceptual model has been constructed, integrating network analysis principles with computational techniques to assess influence. This model introduces a novel framework for evaluating social network nodes based on a combination of structural, semantic, and emotional factors. Specifically, it incorporates influence assessment metrics that consider the heterogeneous nature of content, including its linguistic features, sentiment, and patterns of engagement. Additionally, the model accounts for the emission dynamics of criminal content, allowing for a more precise determination of high-risk nodes within the network. A method for quantifying the influence of social network nodes engaged in criminal content dissemination has been formulated. This method utilizes centrality measures along with content analysis techniques to improve accuracy in detecting key actors. Experimental validation conducted on multiple real-world datasets (including VKontakte groups and known extremist networks) demonstrated that the proposed method achieves an accuracy of up to 80% in identifying the most influential criminal nodes. Compared to baseline centrality-based methods, our approach provides more reliable detection due to the integration of semantic-emotional metrics and emission indicators. The results confirm the practical value of the method in operational scenarios such as the early detection of criminal activity and the prioritization of threat actors for monitoring. These findings have strong implications for real-world applications in law enforcement and cybersecurity. By leveraging advanced algorithmic techniques for social network monitoring, authorities can proactively detect and mitigate the spread of criminal information.

The study focuses on improving the Quality of Service (QoS) in Vehicle-to-Vehicle (V2V) communication within Vehicular Ad Hoc Networks (VANETs) by enhancing the Learning Automata-based Ad Hoc On-Demand Distance Vector (LA-AODV) routing protocol. Unlike the standard AODV, which is a reactive routing protocol, and previous configurations of LA-AODV, this research introduces a fine-tuning strategy for the learning automata parameters. This strategy allows the parameters to dynamically adapt to changing network conditions to reduce routing overhead and enhance transmission stability. Three modified versions of LA-AODV referred to as setups A, B, and C, are evaluated against the standard AODV and earlier LA-AODV configurations. The performance of each setup is measured using key QoS metrics: flood ID, packet loss ratio (PLR), packet delivery ratio (PDR), average throughput, end-to-end delay, and jitter. These metrics are crucial in evaluating the efficiency, reliability, and performance of V2V communication systems within VANETs. The results demonstrate that the LA-AODV variants significantly reduce flood ID counts, which represent the number of times a packet is broadcasted, compared to AODV, with setups A and B achieving reductions of 10.24% and 28.74%, respectively, at 200 transmissions, indicating enhanced scalability. Additionally, LA-AODV setup A provides 5.4% higher throughput in high-density scenarios. The modified versions also significantly decrease delay and jitter, achieving reductions of over 99.99% and 99.93%, respectively, at 50 transmissions. These findings underscore the adaptive capabilities of the proposed LA-AODV modifications, providing reassurance about the robustness of the system. They also highlight the importance of parameter optimization in maintaining reliable V2V communication. Future work will benchmark LA-AODV against other state-of-the-art protocols to validate its effectiveness further.

The paper presents the development of a smart tool for automated analysis of news text content in order to identify propaganda narratives and disinformation. The relevance of the project is due to the growth of the information threat in the context of a hybrid war, in particular in the Ukrainian information space. The proposed solution is implemented in the form of a browser plugin that provides instant analysis of content without the need to switch to third-party services. The methodology is based on the use of modern natural language processing (NLP) and deep learning methods (in particular, BERT models) to classify content according to the level of propaganda impact and identify key narratives. As part of the study, modern models of transformers for text analysis, in particular BERT, were used. For the task of classifying propaganda, pre-trained GloVe vectors optimised for news articles were used, which provided the best results among the options considered. Instead, the BERT model was used to classify narratives, which showed higher accuracy in the processing of texts reflecting subjective thoughts. The adaptation included the use of a multilingual version of BERT (multilingual BERT), as it allows you to effectively work with Ukrainian-language data, which is a key advantage for localised analysis in the context of information warfare. Before using BERT, pre-processing of texts was carried out with the addition of syntactic, punctuation, emotional and stylistic features, which increased the accuracy of classification. For a more complete and reliable assessment of the effectiveness of propaganda classification models and narratives, a set of key metrics was used for propaganda/ narratives analyses Accuracy (0.94/0.86), Precision (0.95/0.69), Recall (0.96/0.71) and F1-score (0.96/0.70).The developed model showed high accuracy results: the F1-score for the propaganda classification problem was 0.96 and for the narrative classification problem – 0.70, which significantly exceeds the results of similar approaches, in particular XGBoost (0.92 and 0.50, respectively). In addition, the system supports full-fledged work with Ukrainian-language content, which is its key competitive advantage. The practical application of the tool covers journalism, fact-checking, analytics, and improving media literacy among citizens, contributing to the improvement of the state's information security.